My dissertation (Hedberg, 1990), undertaken at the University of Minnesota Linguistics Department, with Jeanette K. Gundel and Michael B. Kac as co-advisors, is entitled "Discourse Pragmatics and Cleft Sentences in English". In Chapter 1, I divided the cleft into four structural parts: the cleft pronoun + the clefted constituent + the copula + the cleft clause. In Chapter 2, I sketched my assumptions about discourse pragmatics.
Then in Chapter 3, I identified two primary approaches that had been taken to the structure of cleft sentences previously. The first approach is the 'extraposition analysis', which proposes a syntactic relation between the cleft pronoun and the cleft clause and identifies the clefted constituent with that composite via the copula. The second approach is the 'expletive analysis', which treats the clefted constituent and the copula as expletive, dummy elements, and posits a direct connection between the clefted constituent and the cleft clause. In this chapter I argue that the copula plays a non-expletive role because clefts are like simple copular sentences in exhibiting a specificational/predicational ambiguity. I also argue that in specificational clefts, the complement of the copula must be of the semantic type of entities (type e) or be type-shiftable down to that type from a higher type. I identify a predicate constraint on clefts in Standard English—*It is happy that he is (complement of type <e,t>) and a quantificational constraint on specificational clefts—*It was everyone who came (complement of type <<e,t>,t>).
In Chapter 4, I argued that there is a syntactic connection between the cleft pronoun and the cleft clause, basing my argument on the idea that the cleft pronoun functions pragmatically like a determiner with regard to cognitive status (Gundel, Hedberg and Zacharski 1993), with the cleft clause functioning as the descriptive portion of the resulting definite DP. I rely extensively on th-clefts in making this argument. For example, Was that John Smith who you introduced me to last week?, where that [one]who you introduced me to last week must be familiar to the addressee but need not be activated. Thus, the cleft pronoun plays a non-expletive role. I argue that the cleft pronoun it functions pragmatically as a definite article in the discontinuous definite description, which therefore need only be uniquely identifiable, although it can also have a higher status. In this chapter I also argue that the cleft clause is syntactically a relative clause, and I sketch a biclausal syntactic analysis of the cleft sentence.
In Chapter 5 and Chapter 6, I turn to a corpus of over 700 clefts drawn from a variety of genres (primarily mystery novels, historical narratives, newspaper opinion columns, and spoken conversations), and examine these clefts for their discourse-pragmatic functions. I argue that the cleft clause can either express the topic of the sentence (Chapter 5) or the comment of the sentence (Chapter 6). In topic-clause clefts, the cleft clause can be directly or indirectly activated, and the clefted constituent can be more or less explicitly contrastive depending on how directly activated the cleft clause is. I also propose a class of 'truncated clefts' that lack cleft clauses when the cleft clause material ;is already in the focus of attention, and also a class of 'sentential focus' clefts, which also lack cleft clauses. In comment-clause clefts, the clefted constituent can be activated and the cleft clause is typically only familiar or uniquely identifiable (i.e. presupposed). I argue that the clefted constituent can function as the topic of the cleft sentence, and when it is a continuing topic can be modified with an additive 'focus' particle (also, even). I also analyze vice-versa clefts as comment-clause clefts. Finally, I examine some rhetorical functions of comment-clause clefts in structuring multi-paragraph segments of discourse.
Title-Contents-Acknowledgements